The training and research plans of this second revised K01 application will directly advance the candidate's long term career goal of developing an independent, transdisciplinary line of research that investigates subtypes of risk for alcohol-related problems and of alcohol use disorders based on phenotypic and genotypic information. Career development is sought in statistical genetics (related to hypothesis generation and testing), molecular genetics, and their integration with alcohol studies. These training objectives will be accomplished through coursework, workshops, seminars, and conferences; through extensive mentoring and consulting with senior investigators whose research is directly relevant to this application; and through development of collaborations. These activities will provide the foundation for the research plan, which combines nontraditional quantitative methods with an intermediate phenotype approach to generate conceptually- and empirically-based hypotheses about neurogenetic influences on physiological processes using highly informative yet small scale (N<500) study data. More specifically, the research plan seeks to add a genetic component to two ongoing NIH-supported projects that examine the coordinated function of multiple dynamically-interrelated biological systems and that provide in depth assessment of psychosocial and psychophysiological characteristics related to alcohol use. Genomic DNA is being collected from young adult participants who span the continuum of substance use behaviors and single nucleotide polymorphisms in GABA-A receptor subunit genes will be used as initial genotype targets. A highly sensitive index of heart rate variability - at rest, during cue exposure and acute alcohol challenge - will be used as an initial intermediate phenotype. These initial targets may identify a subset of individuals with high addiction liability due to poor modulation of emotional arousal. The overall aim is to use advanced statistics as a novel approach to identifying the multiple pathways of risk for problematic alcohol use behaviors and suggesting new, targeted approaches to treatment. Hypotheses generated from these discovery-oriented, early-stage studies will subsequently be tested with independent samples gathered through collaborations and in de novo R01 applications. This application applies cutting-edge statistical strategies to the analysis of genetic and physiological data in an effort to explore the genetic basis of drinking behaviors in a new way. This approach may be useful for identifying subtypes of risk for alcohol-related problems and suggesting ways to make prevention, intervention and treatment programs more effective.